The long term goal of our research is to understand the molecular basis of hereditary nonpolyposis colorectal cancer (HNPCC). We previously showed that tumors arising from 90% of HNPCC had microsatellite instability (MSI). Genetic, biochemical, and functional studies in a variety of model systems have demonstrated that MSI is caused by defective cellular DNA mismatch repair (MMR). Human MMR requires six different genes. It is believed that inherited germline defects in four of these genes, hMSH2, hMLH1, hPMS1, and hPMS2, cause HNPCC. To gain a comprehensive understanding of HNPCC, we have analyzed 109 HNPCC kindreds meeting the Amsterdam Criteria for the evidence of an inherited mutation in the entire coding region of the four genes by DNA sequencing. Based on the results of our experiment, we have divided these HNPCC into three different groups. The first group consists of 37 kindreds with defined germline MMR mutations. The second group consists of 24 kindreds with abnormal MMR cDNA deletions or insertions but without detectable genomic mutations. The third group consists of the remaining 48 kindreds that do not show any detectable germline MMR mutations. Nevertheless, MSI was present in tumors arising from all the group II and 37 of the 48 group III kindreds, indicating homozygous mutation in the MMR gene. Our hypothesis is that like group I, MSI positive group II and III HNPCC is also caused by a defective germline MMR gene. To test this hypothesis, we propose to continue our current investigation on group II and III HNPCC using a variety of approaches, including genetic, biochemical, and functional studies as outlined in the following specific aims. AIM 1: To determine if the observed abnormal group II cDNA deletions or insertions segregate with the disease; AIM II: To examine if genomic deletions and nonbonding region mutations of the four MMR genes play any role in these HNPCC; AIM III: To determine if other MMR genes, hMSH3 and hMSH6, are involved in these HNPCC; AIM IV: To determine if MMR gene mutation is present in tumors arising from group II and III HNPCC; and AIM V: To determine if the function of the predisposed germline MMR allele from group II and III HNPCC is inactivated using mono- allelic-based-functional-analysis (MAFA), in which the presumed mutant HNPCC allele is separated from the wild type allele by hybrid fusion with an MMR-null hamster cell line. In combination, these interconnected approaches should lead to new insights into the molecular basis of HNPCC, with important implications for the clinical diagnosis and therapy of the disease.